reduction potential of glutathione - The Journal of Organic

Jul 1, 1993 - Wenjin Yang , Raymond V. Fucini , Bruce T. Fahr , Mike Randal , Kenneth E. Lind , Melissa B. Lam , Wanli Lu , Yafan Lu , Douglas R. Cary...
40 downloads 10 Views 402KB Size
J. Org. Chem. 1993,58, 4144-4146

4144

Oxidation/Reduction Potential of Glutathione

Table I. Half-Cell Potential of the Glutathione/ Glutathiaqe Disulfide ILedor System b

Kevin K. Millis, Kim H. Weaver, and Dallas L. Rabenstein' Department of Chemistry, University of California, Riverside, California 92621 Received September 9,1992 (Revised Manuscript Received April 19, 1993)

Introduction Although the importance of the glutathionelglutathione disulfide (GSH/GSSG) redox system in biological oxidation/reduction reactions has long been recognized,l there still is uncertainty as to its half-cell p ~ t e n t i a l . To ~~ illustrate, literature values ranging from -0.17 to -0.27 V are listed in Table I. Direct measurement of the half-cell potential for the GSHiGSSG system, and other thiol/ disulfide redox systems, by standard electrochemical methods is not possible due to formation of stable metalthiolate complexes at electrode surfaces.8 Thus, the halfcell potentials of GSH and other thiol-containingmolecules are determined indirectly by measurement of equilibrium constants for their reaction with redox systems of known half-cell potential? The first three values and the fifth value listed in Table I for E&G/GsH were determined indirectly by measurement of the equilibrium constant for reaction of the GSHI GSSG redox system with the nicotinamide adenine dinucleotide phosphate (NADPH/NADP+) redox system."ve The fourth and sixth values were determined

GSSG + NADPH + H+ F;? 2GSH + NADP'

(1)

indirectly relative to the nicotinamideadeninedinucleotide (NADH/NAD+)redox s y ~ t e m . ~The 9 ~ large range covered by the values listed in Table I is probably due in part at least to limitations of the experimental procedures used, since equilibrium constants were determined by direct measurement of the concentration of only one or two of the species involved in the equilibrium reaction^.^^ In view of the fundamental importance of the GSHI GSSG redox system in biology, and ita use as the reference system for determination of redox potentials for other thiolldisulfide systems, we have redetermined E&sG/GsH relative to the half-cell potential for the (1) Meistar, A. In Glutathione: Chemical, Biochemical and Medical Aspects; Dolphin, D., Pouleon, R., Avramovic, O., E&.; Wiley-Interscience: New York, 1989; Part A, pp 1-48. (2) Rall, T. W.; Lehninger, A. L. J. Bid. Chem. 1952, 194, 119-130. WBB corrected The value reported by Rall and Lehniiger for &mlosH by Roet and Rapoport6 using a more accurate value for Ed ADPH' (3) Mapson, L. W.; Isherwood, F. A. Biochem. J. 1 9 6 3 , w iy3-191. was calculated by Roet and The value lieted in Table I for Ed and Isherwood for the Rapoport6 wing the value repo&&%apen equilibrium constant. (4) Scott, E. M.;Duncan, I. W.; Ekstrand, V. J. Biol. Chem. 1963,238, 3928-3935. - .- - - - - -. (5) Roet, J.; Rapoport, S.Nature 1964,201, 186. (6) Veech, R. L.; Eggleeton, L. V.; Krebs, H. A. Biochem.J. 1969,115, was calculated w609-619. The value listed in Table I for Ed ing the value reported by Veech et al. for thG%%rium constant and N p W p H ~ 4 . 3 2 4V. #qpdci, R. P.; Whitaides, G. M. J. Am. Chem. SOC.1980,102, 2011-2026. (8) Jocelyn, P. C. Biochemistry of the SH Group; Academic Press: New York, 1972; p 56.

&fi

condne

ELIMH

KO

1.2 x loae 7.0 X lW

-0.17 -0.16

1.55 X 1 P

-0.25

9.8 X 1ozd

-0.24

51'

-0.28

6.7 X losd

1.39(*0.21) X lozd

-0,205 -0.263a

1.20(*0.15) X 1ozd

-0.26P -0.252

pH7.0 pH 7.4,20 "C, 0.1 M phosphate pH 6 . 8 , s "C, 0.167 M phosphate pH 7.4,40 O C , 0.125 M phosphate pH 7.0,38 OC, Z = 0.25 M pH 7.0,30 OC pH 7.07,25 OC, 0.10 M phosphate pD 6 . 9 7 , s OC, 0.10 M phoephate pD 7.0,O.lO M phosphate

ref 2 3 4

5 6 7 this

work thie work 17

'Conditional equilibrium constant ( ~ 8defied by eq 2 or the analogow equilibrium constant for the reactionwith NADH. b Volta va the standard hydrogen electrode. 0 Equilibrium constant for the reaction given in eq 1. d J3quilibrium constant for the reaction of GSH/GSSG with the "+/NADPH redox system. Calculated wing EE;ADpINADpA = -0.324 V at pH 7.0 and 25 "C. See ref 11 for values calculated using a different value for E;mpmmpH.

NADPH/NADP+ redox system by lH nuclear magnetic resonance (NMR) spectroscopy. Compared to the spectrophotometric methods used p r e v i o u ~ l ylH , ~ ~NMR has the advantage that the concentrations of all four species involved in the equilibrium reaction can be determined directly. Values are reported for E&GIGSH a t pH (pD) 7.0 in HzO and DzO solutions.

Results The concentrations of GSH, GSSG, NADP+, and NADPH in equilibrium solutions were determined by lH NMR. To illustrate, the 'H NMR spectrum of an equilibrium mixture prepared by reaction of 60 mM GSH with 30 mM NADP+ in 0.1 M phosphate buffered HzO solution, pH 7.07, 25 "C, is presented in Figure 1. Glutathione reductase was added to catalyze the reaction? It is clear from the spectrum that the equilibrium for the reaction given by eq 1 lies far to the right, with the formation of only small amounts of GSSG and NADPH. The resonances used to determine the equilibrium concentrations of GSH, GSSG, NADP+, and NADPH are identified in Figure 1 and assigned in the figure legend. The equilibrium constant (eq 2) calculated using these

Kc =

[GSH12[NADP+] [GSSG] [NADPH]

concentrations is a conditional constant, as indicated by the subscript, which incorporates the pH (pD) of the was solution. The formal half-cell potential, calculated from the equilibrium constant using eq 3. An average value of 139 f 21 M was determined from 17 measurements (n = 17) of the equilibrium constant in HzO solution a t pH 7.07 and 25 "C. A value of 4 . 2 6 3 f (9) Floh6, L.; Ghzler, W. A. In Glutathione: Metabolism and Function; Arias, I. M., Jakoby, W. B., Eds.; Raven Press: New York, 1978; pp 17-34.

0022-3263/93/1968-4144$04.00/0 0 1993 American Chemical Society

Notes

J. Org. Chem., Vol. 58,No.15, 1993 4145 oz’

m

1

n77l-rn 3.34

3.28 ppm

llllllllllrrlll1l 6 96 6.91 ppln

N2’

I\ /

9

~

~

8



A ~

7

,

~

6

~

~

5

~

4

,

~

3

~

~

2

~

l

w m

Figure 1. The 5OO-MHzlH NMR spectrum of an equilibriumreaction mixture of NADPH, NADP+,GSH, and GSSG in 0.1 M aqueous phosphate solution, pH 7.07, 25 O C . The initial concentrations of GSH and NADP+ were 60 and 30 mM, respectively. The H20 resonance at 4.77 ppm was suppressed by selective presaturetion for 9 s prior to the nonselective observation pulse. The resonances used to obtain equilibrium concentrations are identified; N2 and N2’ are resonances for the proton on C2 of the nicotinamide ring of NADPH (N2) and NADP+(N2’), g2‘ is one half of the AB part of the ABX pattern for the &CH2 protons of the cysteinyl residues of GSSG, and g3, g3’ and g4, g4’ are for the 6-CH2and 8-CH2 protons, respectively, of the glutamyl residues of GSH and GSSG. The expansions of the g2’ and N2 resonances are plotted with a vertical scale 50X that was used to plot the entire spectrum.

0.002 V is calculated for using this equilibrium constant and a value of -0.326V for the reference potential Eo&p/NmpH.1G12 An average value of 120 f 15 M (n = 29)was determined for the equilibrium constant in D20 solution at pD 6.97and 25 OC, from which a value of -0.262 f 0.001 V was calculated for EO&,G9H using a value of -0.323V for Egmp,Nmpw13 The uncertainties of fO.OO1 or f0.002V represent the precision of the measurements; however, the actual uncertainty in the E L l G s Hvalues may be as large as fO.01 V due to uncertainty in the 11 reference value used for

Discussion The values determined in this work for E O d m / G s H are based on measurement of the equilibrium constant for the reaction given in eq 1. The first five values listed in Table I were also calculated from equilibrium constants for this reaction’*6 or the analogous reaction with the NADH/NAD+ redox systemas However, the concentrations of only one or two of the species involved in the -0.326 v VB the standard hydrogen (10) The Value Of h&p/”p~ electrode at pH 7.07 was calculatedfrom Ed = -0.324 V at pH 7.0 and 25 OCll using a pH dependence o??%$’#?V /pH unit. (11) Several valuw have been reported for$ p,NADpH (Clark, W. M. Oxidation-Reduction Potentiale of Organic &stem; Williams and Wilkins: Baltimore, 1960; pp 495-496). The value of Ed -0.324 V is the most widely used; however, accordi&%t%c $ -0.317 0.002 V at pH 7.0 and 30 O C (reported by: &ey,Y. Donovan, J. A. J. Biol. Chem. 1969,234,677-880) might be more accurate. Using the value reported by Rodkey and Donovan, a temperature dependence of -0.0013 V/OC, a pH dependence of -0.0301 V/DHunit,and the eauilibriumconstantdetermined in thiswork. calcuiatad to be -0.250 0.002 V at pH 7.07 and 25 OC. . Ed%!%%dkev. F. L. J. Biol. Chem. 1955,213,777-786. (13) The vkue of EE);AD~/NDPH= -0.323 v vs the standard hydrogen electrode at pD 6.97 was calculated from E&,pINADpH = -0.324 V at pH 7.0 and 25 OC since no values have been reported for Ed tgDP/NADpfi’” DsO solution. As diecussed in ref 11, a less commonly wed ut poeei y more accurate value for is -0.317 V at pH 7.0 and 30 ‘C. Using this value, a temperature dependence of -0.0013 V/’C and a pH dependence of -0.0301 V/pH unit, EdassolosH in pD 6.97 solution is calculated to be -0.250 0.001 V.

1.;

is

*

*

equilibrium reaction could be determined directly with the experimental methods used to determine the first five equilibrium constants in Table I. For example, the first two equilibrium constants were determined by measuring the concentration of NADPH by spectrophotometry.2J The equilibrium concentration of GSSG was then assumed to equal the concentration of NADPH produced by the reaction, and the concentrations of GSH and NADP+ were calculated from these concentrations and the initial concentrations. Since the equilibrium for the reaction (eq 1) lies far to the right (Figure 11, the equilibrium concentration of GSSG produced by the reaction of GSH with NADP+ is small, and thus the values calculated for H be in error the equilibrium constant and E o ~ ~ G / G swould if there were appreciable amountsof GSSG initially present in the GSH.I4 The third and fourth values listed in Table I were also determined spectrophotometrically; however, experimental procedures were used which would take into account any GSSG initially present in the GSHe4eSThe fifth value reported in Table I was also determined by measuring the concentration of NADPH by spectrophotometry; however, details of the experimental procedure were not reported.6 The sixth value in Table I was obtained indirectly by a procedure which involved determination of the equilibrium position of two simultaneous redox reactions: the reaction of the GSH/GSSG redox system with the reduced lipoamide/oxidized lipoamide (lipded/lipoox)redox system and the reaction of the lip@/lipooX redox system with the NADH/NAD+ redox system.’ The concentration of NADH was determined by spectrophotometry, and then the concentrations of the five other species in the two simultaneous equilibria were calculated using this concentration. Calculation of E&GlGSH required the use of literature values for both the lipord/lipooxand NADH/ NAD+ redox potentials. Although direct comparisons are not possible due to small differences in solution conditions, the values de(14) We typically find up to 0.3% GSSG in the GSH; others report 0.5-1.0% GSSG.6

~

~

4146

J. Org. Chem., Vol. 58, No.15, 1993

Notes

Table 11. Half-Cell Potentials of ThioVDisulfide Redox Systemsthiol EkWRSH* thiol captopril -0,287 i 0.001 homocysteine NJV-dimethyl- -0.271 f 0.003 8-mercaptocysteamine ethanol

E b H P

-0.256 i 0.005 -0.253 f 0.003

glutathione cysteamine

*

-0.262 i 0.001 cysteine -0.245 0.001 -0.260 i 0.001 penicillamine -0.243 3-mercaptopro- -0.257 f 0.004 2-amino-2-methyl- -0.221 i 0.002 pionic acid 1-propanethiol coenzyme A -0.256 f 0.003 0 Relative to the standard hydrogen electrode. Calculated using ElmlqH.= -0.262 V. pD 7.0,26 "C. The uncertainties represent the precision of the measurement;the actual uncertainty may be as lafge as iO.01 V due to uncertainty in the reference value for ENADPINADPH, as discussed in the text.

termined in the present work for EOiMG,aH are in good agreement with those reported by Scott et al.' and by Rost and R a p o p ~ r t .As ~ discussed above, these would also appear to be the most reliable values listed in Table I since they were determined by procedures which would account for any GSSG initially present in the GSH. Recently, half-cell potentials were determined for a group of thiol-containing biological molecules and related molecules by measuring equilibrium constants for their thiol/disulfide exchange reactions with, in most cases, the GSH/GSSG system.15 Values were calculated for 2RSH + GSSG * RSSR + 2GSH

(4)

E&sRIR9H from the equilibrium constants using the value reported by Szajewski and Whitesides' for Eo&sol~wWe have recalculated the half-cell potentials for these thiolcontaining molecules relative to the value determined in this work for E&SG/GsH. The resulta are reported in Table 11.

Experimental Section

Solutions were prepared in sodium phosphate buffered H20 or D2O solution (0.125M buffer; pH 7.0 or pD 7.0). Standard solutions of GSH and NADP+ were prepared in a nitrogen atmoaphere in a Plas Labs, Inc. Mode1885-AC anaerobic chamber or a glovebag. The nitrogen gas was passed through a column of BASF copper catalyst R3-11to remove traces of oxygen. The concentrations of the standard solutions, including the concentration of GSSG initially present in the GSH solutions, were determined by 1H NMR, using glycine as an intemal intensity standard. On the basis of the concentrations determined for the standard solutions,appropriate volumes of the standard solutions together with 7-10 units of glutathione reductase were combined directly in NMR tubes in the anaerobic chamber or the glovebag. In some cases, glycine was added to the reaction mixture as an intemal intensity standard and the equilibrium concentptions were determined from the relative resonance inknsities. The NMR tubes were capped, the caps were wrapped with parafilm to further exclude oxygen, and then the tubes were covered 'to minimize light-induced degradation of NADPH. The p I j or pD of the solutions was measured directly in the NMR'tubes in @e anaerobic chamber or glovebag with an Ingold microelectrode, which was calibrated usingpH 4.00and 7.00aqueous buffers. pH meter readings for D20 solutions were converted to pD values using the equation pD = pH + 0.40. The initial concentrations of GSH and NADP+ used in the experiments conducted in HzO solution were, respectively: 60 and 30 mM (n = 4),38 and 19 mM (n = 11, -85 anq 12 mM (n = 6), and -42 and -30 mM (n = 6). Those used in the experimenta conducted in D20 were, respectively: 42 and 17 mM (n = 2),53and 20 mM (n= 31, -80 and -10 mM (n= 121, -38 and -25 mM (n = 61, and -44 and -20 m l (n= 6). To establish that equilibrium was achieved before NMR spectra were measured, spectra were measured as a function of time for 200 min after the GSH, NADP+, and enzyme solutions were combined in two separate experimenta. It was found that equilibrium had been achieved by the time the fist spectrum was measured (20min after mixing). All spectra wed to calculate equilibrium constants were measured at least 30 min after the r e a h t a were combined. To establish that the same equilibrium cons@nt is obtained when equilibrium is approached from the other direction, NMR spectra were measured for equilibrium solutions prepared by combining standard solutions of GSH, NADP+,and enzyme. Then GSSG standard solution was added to the NMR tube to increaee the GSSG concentration bya factor of two to four times the equilibrium concentration, and the system was allowed to return to equilibrium by the reverse reaction. Values of 145 f 15 (n = 6) and 142 i30 (n = 6)were obtained for the equilibrium constant in H20 from spectram&ed before and after the addition of GSSG. The corresponding values for D2O solution are 127 f 17 (n = 12) and 117 f 13 fp = 12).

NADP+,glutathione, and glutathione reductase were obtained from Sigma Chemical Co. glycine from Aldrich Chemical Co., and D20, DC1 (35% in DzO), and NaOD (40% in DzO) from Isotech Inc. lH NMR measurements were made at 500 MHz. Ti values were estimated by the inversion-recovery method for the analytical resonances (Figure l) used to quantitate the equilibrium mixtures using separate solutions of GSH, GSSG, NADPH, and NADP+. 1H NMR spectra were measured for the equilibrium mixtures using relaxation delays 4.5 times the longest TI of the analytical resonances together with a flip angle of 84". The HOD or Ha0 resonances were suppressed by selective presaturation. Relative resonance intensities for the analytical resonances were determined by the cut and weigh methodi@or with the standard software on the Varian VXR-600s spectrometer.

Acknowledgment. This research was supported in part by National Institutes of Health Grant GM 37000 and by the University of California,Riverside, Committee on Research. The NMR instrumentation was supported in part by BRSG 2 S07RR0701G20awarded by Biomedical Research Resources, National Institutes of Wealth.

(15) Keire, D. A., Straw, E.; Guo, W.; Nos&, B.; Rabenatein, D. L. J. Org. Chem. 1992,67, 123-127. (16)Harris, W. E.;Kratochvil, B. Chemical Separations and Mewurements: Background and Procedures for Modern Analysis; W. B. Saundera Golden Series: Philadelphia, 1974; p 197.

(17) After this paper wae submitted, a value of -0.252 V wan reported by: Lees, W. J.; Whitesides, G. M.J. Org. Chem. ISSS,MI, 642-647. Thia value WBB detarmined relative to the Eo' value for lipoic acid using equilibrium constants measured by N M R for the lipoic acid/oxiW mercaptoethanol and mercaptoethanoYglutathionedisulfide redox reactions.